P2P-VOD缓存策略MVC

本文介绍了P2P-VOD系统中的MVC缓存策略,该策略是为了改进流媒体系统视频播放的流畅程度,服务器负载压力而提出的。P2P-VOD系统视频的连续播放,服务器的被请求频率和用户的满意程度是系统重要因素,然而现在系统中的SVC缓存策略没有充分考虑到冷热门频道节点重叠存储数据的情况,由此本文提出MVC缓存策略,该策略充分考虑到点播系统中视频冷热度,流行值等会引起系统整体缓存不均衡的因素,提高冷门视频数据分块缓存量,在节点多的情况下,热门频道节点帮助冷门频道下载数据分片,在服务器通信量不足的情况下,提高视频播放的流畅程度,减轻服务器的负载压力。     

一种基于节点可选度的P2P点播流媒体数据调度算法

当前的P2P(Peer-to-Peer)点播流媒体系统中数据调度算法未能充分利用每个用户节点自身的特性.在分析典型数据调度算法基础上提出一种基于节点可选度的数据调度算法(SSP算法).该算法一方面在调度下载数据块时综合考虑了邻居节点带宽能力及其所拥有的数据信息.另一方面对服务节点的请求处理过程进行了优化.SSP算法有利于提高用户节点播放视频的连续性,降低流媒体服务器的负载压力,从而改善P2P点播流媒体系统的整体服务质量.仿真结果和实际应用表明算法性能良好,适用于用户节点能力差异较大的P2P点播流媒体环境.     

一种带宽自适应P2P视频点播数据调度策略

在对等网络视频点播系统中,节点的服务能力差异较大,充分利用各类节点的服务能力是一个值得研究的问题。数据请求量是请求周期中向各个邻居节点请求数据包的数量,在数据调度中是一个重要的因素。为此,从数据请求量的角度出发,提出一种带宽自适应数据调度请求策略,当邻居节点具有剩余带宽时,本地节点根据自身的贡献潜力自适应增加相应的请求量,加速数据片段在高带宽节点中的扩散速度,让高性能节点更快地具备更多的内容为其他节点提供服务,从而降低服务器负载压力。仿真实验结果表明,该策略能进一步提高P2P点播系统的整体性能。     

基于仿真的P2P直播系统的研究

随着互联网技术的不断发展,基于P2P技术的网络视频直播和点播系统受到广大科研人员和用户的关注。与传统的C/S结构的视频播放系统相比,P2P直播系统可以用户提供更加流畅、更加清晰的视频服务。本文分析P2P直播系统的参数和工作原理,提供同构网中服务器负载的分析。     

基于遗传算法的P2P流媒体数据调度策略研究

介绍一种基于P2P的点播系统中的数据调度策略。该策略的主要特点是使用了遗传算法进行寻优,并针对该问题提出了相应的编码方式、适应度函数,罚函数以及控制服务器负载平衡的修补策略,在服务器和P2P网络中节点之间进行合理的分配调度数据,从而充分利用P2P网络中节点的资源以减轻服务器的负载。最后通过仿真实验,验证了算法的有效性。     

基于负载均衡的多源流媒体网拓扑优化策略

分析多源流媒体网上下层拓扑,提出一种基于负载均衡的拓扑优化策略。该策略充分利用网络资源,结合媒体分发服务器功能,采用P2P技术和组播原理,实现多源流媒体网的拓扑均衡以及媒体服务器和网上所有节点的负载均衡,在保证系统可扩充性的基础上,降低播放延时并增强系统稳定性。     

OXVoD：一个基于分层结构的P2P视频点播系统

可扩展性和高播放连续度是视频点播系统大规模应用的关键。提出了一个分层结构的P2P点播系统,融合了Distributed Hash Table（DHT）的精确高效和Gossip协议的简单实用。该系统上层结点为下层结点提供下载服务,有效均衡负载。提出了一种数据调度策略,通过在基于Gossip协议数据可用信息交互过程中添加少量本地信息,帮助下载者选择最优下载源,提高系统播放连续度。仿真实验表明,OXVoD可以在服务器负载稳定的情况下,保证99%以上的结点正常加入系统,并获得96%以上的播放连续度。     

P2P流媒体点播系统中的缓存管理方案的研究和实现

视频点播系统的缓存管理对系统的性能有重要的影响.回顾了现有的缓存策略,分析了P2P网络和传统中心服务器网络缓存策略的不同,然后提出一种适应P2P流媒体点播系统的缓存管理算法.给出了新型的算法描述,并构建原型系统进行性能对比测试.测试结果表明系统骨干节点的资源利用率和普通节点节目的播放质量和流畅性都得到了提高,证实了算法在P2P点播系统中的可行性和有效性.     

P2P异步传输点播系统设计与仿真

互联网上大规模Peer-to-Peer(P2P)点播服务存在高效资源调度和不可预知用户行为等挑战.借鉴BitTor-rent(BT)算法原理,提出基于异步传输的P2P点播系统.该系统采用集中式节点定位以及非结构化拓扑,通过异步传榆机制提升文件资源共享度和利用率,并引入视频数据块的优先级和稀有程度改进系统播放连续度.数据包及仿真实验表明本系统在播放连续度、启动时延以及服务器压力等性能指标上的良好性能,充分验证了点播系统中异步传榆机制的可行性.     

基于P2P的流媒体客户端缓存管理机制的研究

针对在基于P2P的点播系统中,由于客户端缓存区没有得到高效的利用而影响流媒体点播系统的服务质量问题,提出了一种新的基于混合P2P的流媒体点播模型P2P_VOD,该模型将客户端缓存分为三个区,并详细阐述了客户端节点缓存区的缓存替换机制,综合考虑了数据块备份量的均衡性和节点VCR操作的命中率,使得节目数据块在各节点间缓存得到全局优化并有效缓解了服务器负载。通过仿真对比实验,验证了该模型在启动延迟和服务器负载方面的优越性。     

基于P2VoD的视频点播系统

基于P2VoD协议,设计了一种能使得节点快速有效地加入网络的VoD系统。考虑P2P网络的动态性,在节点的加入过程中为失效恢复预留带宽资源。将节点的缓存分为代内缓存和代外缓存,充分利用节点的缓存提高播放的流畅性。带宽不足节点及时向服务器报告带宽信息,减少节点加入时被拒绝的概率。仿真结果表明,与原有系统相比,新系统在充分利用服务器负载能力的情况下,明显地降低了节点加入系统的延迟,增加了节点加入系统的概率,并能够保证节目播放的连续性。     

一种基于P2P的短视频分享系统

现有在线短视频分享策略通常采用C/S架构,给视频服务器带来较大的带宽压力。为此,提出一种采用点对点方式的在线短视频分享系统IShare,该系统结合用户点播偏好和视频文件之间的社会网络特性实现视频分享。IShare主要包括基于点播兴趣的节点分簇和视频数据源节点的查找2个核心技术。实验结果表明,IShare具备较好的视频数据源节点查找能力,可降低视频服务器带宽资源消耗。     

Integrated buffering schemes for P2P VoD services

How to improve the scalability and QoS of peer-to-peer on-demand streaming system based on unstructured overlay is still a problem. Researchers have proposed some memory based buffering schemes to archive the targets. Considering the limited space of memory on one peer, a new caching strategy, which can integrate memory-caching strategy with disk-caching strategy, is proposed to make full use of peers memory, disk and bandwidth resources. Based on the new strategy, peers can request media data from neighbors of the overlay, buffer the fresh part into the memory slots and the watched part into the free local disk, which can enlarge the capacity to buffer media data. Based on the new scheme, the experimental results show that the new caching strategy improves the service capacity and QoS of the whole system greatly. The load of the media server is obviously alleviated and the continuity of playing media data is obviously improved.     

Modeling and analysis of multi-channel P2P VoD systems

Peer-to-Peer (P2P) technology has recently become a tremendously attractive solution to offload servers in large-scale Video-on-Demand (VoD) applications by utilizing the upload capability of participating peers. However, the alleviation of server load may not be very effective in the multi-channel P2P VoD systems due to the intra-channel and inter-channel upload bandwidth imbalances. In this paper, we evaluate the impact of these bandwidth imbalances on the system performance through two analytically models: a queuing network model which describes the whole system, and a stage-based model which describes a single channel. Our models capture several aspects of peer behavior, such as participating in the system, sojourning in a channel, downloading and uploading the content, wandering around channels and leaving the system. We apply our models to three potential P2P VoD designs: Naive bandwidth Allocation Design (NAD), Independent-channel Chunk-aware bandwidth Allocation design (ICA) and Cross-channel Chunk-aware bandwidth Allocation design (CCA). By developing an asymptotic theory to provide theoretical results, we analytically show that ICA can perform better than NAD, and CCA can perform best of the three designs in terms of server load. We also present optimization problems and simple heuristic strategies for ICA and CCA. Our analytical results are validated by extensive simulations.     

Segmented patching broadcasting protocol for video data

In adaptive segment-based patching scheme, the video is divided into fixed number of segments, which are transmitted over the server channels. For efficient transmission of the video segments, the server channels are classified into two types – regular and patching channels. A regular channel generally transmits fixed number of segments and a patching channel helps transmitting those segments that cannot be provided by any regular channel to the users. The number of segments transmitted by the first regular channel is decided by the number of regular channels that are allocated to the video by the video server. Other regular channels transmit pre-specified number of segments. This scheme estimates the bandwidth of the patching channels based on the requests received at the video server in terms of fixed time intervals, called time slots. The bandwidth estimation in this scheme is less accurate because for multiple requests received in a time slot more than one patching channels are used. Second, the probability distribution considered in this scheme does not satisfy the basic rule, i.e., the sum of all probabilities is not 1. In this paper, we address these issues and propose a new protocol named as Segmented Patching Broadcasting Protocol for Video Data. The average server bandwidth allocated to the patching channels is much less as compared to the adaptive segment-based patching scheme because only one patching channel is sufficient for any number of requests received in a time slot.     

P2P VOD系统邻居节点查找算法

P2P流媒体点播系统中,由于用户的VCR操作存在不确定性,在查找合作节点时存在一定困难。针对该问题,提出了一种基于预测的邻居节点选择算法。描述了基于虚拟直播频道的点播系统覆盖网构建方式,将系统中的节点通过DHT协议的方式组织起来。提出了一种基于用户行为的学习预测策略模型,根据该模型预测结果提前查找所需邻居节点。仿真实验表明,该算法能改善用户跳转响应延时,促进节点协作。     

基于P2P视频点播技术的流媒体平台设计与开发

随着流媒体应用在Internet上的流行,传统C/S模式的流媒体服务系统已经不能满足流媒体对服务器性能和高带宽的要求,严重阻碍了流媒体业务质量的提高和容量的扩大。本文介绍一种基于P2P网络的流媒体播放技术,它将P2P网络技术和流媒体技术结合起来,充分利用客户计算机的资源,减轻流媒体服务器和网络负载,突破了传统的流媒体播放系统带宽瓶颈,能够保持播放节目流完整而流畅。本文还采用MVC模式和Java语言以面向对象方法设计和开发P2P流媒体网站,利用P2P流媒体技术,实现校园流媒体的视频点播。     

基于P2P的用户生产内容视频资源查找策略

现有用户生产内容(UGC)类视频系统通常采用C/S架构设计,导致了视频服务器极大的带宽压力。提出一种采用对等网(P2P)的在线短视频查找策略——FastSearch,其目的是利用视频资源之间的关联关系进行视频资源定位,以显著提高点播节点之间的视频分享效率并降低对视频服务器的带宽需求。实验表明FastSearch具备良好的视频数据源节点查找能力,集成了该查找策略的短视频系统能有效减少对视频服务器的带宽消耗。     

An efficient server bandwidth costs decreased mechanism towards mobile devices in cloud-assisted P2P-VoD system

Recently, more and more devices with small buffer size such as PDAs or mobile phones are joining in the VoD system, which leads to two major challenges: how to efficiently distribute their bandwidth resources with small buffer size, and how to provide assistant mechanism to make them playback smoothness. In face of this situation and for the purpose of decreasing the server bandwidth costs, we propose a peers’ downloading mechanism called NCDLT to solve above challenges. It contains two algorithms. The first is neighbors and chunks downloading selection (NCS) algorithm and it ensures peers to find neighbors who can provide video data with lower refusal rate. The second is distributed linear taxation algorithm (DLT) and it makes peers with lower capability acquire enough download rate to reduce the request to servers. The simulation results demonstrate that our algorithms can offload the server bandwidth costs and improve the download rate of peers with small buffer size.     

Video Data Delivery using Slotted Patching

In the request-based scheme, a patching channel is initiated on arrival of a request in an interarrival time of two adjacent multicast channels. If two requests are received in a short duration in an interarrival time, two patching channels are needed to service them. Both these channels deliver almost the same data, thus, wasting the bandwidth. This problem may be addressed in two ways: using higher level patching technique, or forcing the first request to wait for the second one. Using higher level patching technique makes the system more complex and thus usually not preferred. In the second case, the bandwidth usage certainly reduces, but the length of the waiting time by the first request for the second request is the main issue. In this paper, this issue is addressed by dividing the video duration into uniform time slots such that there is at least one request in a time slot for initiating the patching channel. The proposed scheme is named as Video Data Delivery using Slotted Patching. In this scheme, the patching channels download much less video data than the request-based scheme and thus require less bandwidth. This scheme may not provide immediate services to all users unlike the request-based scheme, but the user's waiting can be made arbitrarily small without using much resources unlike the request-based scheme.